Excessive adiposity is emerging as the nation's most significant health problem. Although total body adipose tissue (AT) accounts for more than one-fourth of body mass in the average American adult, only certain relatively small highly active compartments are strongly associated with morbidity. The most widely studied of these is visceral adipose tissue (VAT). Measuring VAT is currently central to phenotyping human subjects for obesity and the accepted reference measurement methods are magnetic resonance imaging (MRI) and computed tomography (CT). CT is of limited applicability due to radiation exposure. The traditional imaging approach for quantifying VAT is to measure the cross-sectional VAT "area" at the level of the L4-L5 vertebrae using CT or MRI. The L4-L5 level is an anatomic location established largely by convenience as representative of total VAT. An extensive recent review completed by our group reveals large discrepancies across studies in VAT measurement by imaging methods. This lack of consensus is troubling in view of the critical need for improved phenotyping and for monitoring small but metabolically significant effects of interventions on important adipose tissue compartments. Recent studies suggest that intraperitoneal AT, mainly omental and mesenteric AT, is the VAT component that most strongly links adiposity with adverse biological outcomes and morbidity. Intraperitoneal AT is not localized mainly at L4-L5, but at higher levels, and we have preliminary data suggesting other anatomic locations may be more strongly related to undesirable levels of metabolic measures. Moreover, total VAT includes multiple AT components and may not be the best marker of intraperitoneal AT or outcomes. To investigate these questions we propose to assemble and analyze the largest-to-date (~3000 scans) and most subject-diverse whole-body MRI database linked to associated body composition and laboratory studies. The pooled images and data will provide the needed initial data and rationale for expanding MRI studies into the emerging area of high-resolution image acquisition. It will also permit testing many other important hypotheses related to methods, model development and biological interests. Our findings could help to set new standards for VAT and adipose tissue subcompartment measurement, lower the cost of MRI body composition studies, and improve study power. [unreadable] [unreadable]